Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 3224
    Prev
    Next
    INDEX
    Explanations

    code structure elements, specifically class definitions and access modifiers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Immediately precedes access modifiers ("public", "protected")

    np_acts-logits-general · gemini-2.0-flash

    agent appointed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Geplaatst
    -0.67
    oredCriteria
    -0.66
    DoubleQuotes
    -0.59
    simpleType
    -0.57
    🏻
    -0.56
     asn
    -0.54
     '\\;'
    -0.54
     кӀ
    -0.54
    PerformLayout
    -0.53
    ametric
    -0.53
    POSITIVE LOGITS
    private
    0.72
    OrBuilder
    0.63
    public
    0.54
     private
    0.54
    ,:),
    0.54
     indisponible
    0.52
    protected
    0.52
     bool
    0.49
    twimg
    0.49
    string
    0.48
    Activations Density 0.034%

    No Known Activations